电力系统经济调度模型约束处理方法研究

论文总字数：23451字

摘 要

近年来，随着时代的发展，各行各业对电能的依赖度不断拔高，电能的需求量也不断增长。但是在当今的市场环境中，电力系统的发展遇上了诸多困难。传统中依托于一次能源发电的系统，面临着能源短缺、环境恶化、发电成本急剧增加的困境。同时，随着电力系统规模的不断扩大，电网中的问题也不断浮现，出现诸如负荷率降低、负荷谷峰差的不断增大的现象。为了节约能源，保证其运行质量，因此在电力系统优化运行方面继续深入研究十分有必要。经济调度是系统优化的核心，其内核是基于调度期间的负荷预测，确定此期间的电网机组发电计划。该计划应在满足所有约束的条件下，使系统消耗最小，运行方式最为经济。约束条件的处理方式直接影响到优化算法求解效率和调度结果。因此，将重点放在约束处理过程中，对系统经济调度问题进行优化求解，具有重要意义。针对上述问题，本文开展了如下工作：

1. 针对电力系统经济调度问题建立的数学模型。对模型的目标函数以及常见的约束条件进行分析说明。介绍了几种常见的传统约束的处理方法，对粒子群算法的原理、模型及求解流程进行介绍，为下文的研究提供理论基础。
2. 采用耗量微增率对等式约束进行分摊的方法，对不带阀点的电力系统约束进行处理。以系统总功率上调或下调的情况为根据，对修正等式约束的不平衡量和各机组出力进行修正，对等式约束的不平衡量进行分摊。当系统总功率上调时，优先上调耗量微增率小的机组；当系统总功率下降时，优先下调耗量微增率大的机组。分别对3机组和6机组两个算例进行仿真求解，对本方法的有效性、可行性进行了验证。并通过与粒子群算法相结合，验证其普适性。并通过两个算例的仿真结果，说明其在求解时间和效率方面，具有一定的优越性。
3. 对于考虑阀点效应的电力系统，依旧采取以耗量微增率为根据，对等式约束的违反量进行分摊，然后与粒子群算法相结合的约束处理方法。通过对3机组和6机组的算例分析，对其求解时间和效率方面的优越性提供说明，同时证明了其寻优结果的稳定性。综合来看，该方法在解决考虑阀点效应的经济调度问题方面，求解结果十分有效，在应用方面有一定的优势。

关键词：经济调度；约束处理；耗量微增率；阀点效应；出力分摊；粒子群算法

Abstract

In recent years, with the development of the times, all walks of life have continuously increased their dependence on electrical energy, and the demand for electrical energy has also continued to grow. However, in today"s market environment, the development of power systems has encountered many difficulties. Traditionally, systems that rely on primary energy for power generation face the dilemma of energy shortages, environmental degradation, and sharp increases in power generation costs. At the same time, as the scale of the power system continues to expand, problems in the power grid continue to emerge, such as the decrease of load rate and the continuous increase of load valley-to-peak difference. In order to save energy and ensure the quality of its operation, it is necessary to continue in-depth research on the optimization of power system operation. Economic dispatch is the core of system optimization, and its kernel is based on the load forecast during the dispatch period to determine the power generation plan of the grid units during this period. The plan should meet all constraints, minimize system consumption and operate in the most economical manner. The way the constraint conditions are handled directly affects the efficiency of the optimization algorithm and the scheduling results. Therefore, it is of great significance to focus on the process of constraint processing and optimize the solution of system economic dispatching problems. In response to the above problems, this article has carried out the following work:

1. A mathematical model established for the economic dispatch of the power system. Analyze and explain the objective function of the model and common constraints. Several common traditional constraint processing methods are introduced, and the principle, model and solution process of particle swarm optimization are introduced to provide a theoretical basis for the following research.
2. The method of apportioning the equational constraints with the consumption micro-increasing rate is used to deal with the power system constraints without valve points. Based on the increase or decrease of the total power of the system, the unbalance constrained by the correction equation and the output of each unit are corrected, and the unbalance constrained by the equation is apportioned. When the total power of the system is adjusted upwards, the units with a small increase in consumption will be given priority; when the total power of the system decreases, the units with a small increase in consumption will be given priority to decrease. Two calculation examples of unit 3 and unit 6 were simulated to verify the effectiveness and feasibility of this method. And through the combination with particle swarm algorithm to verify its universality. The simulation results of two calculation examples show that it has certain advantages in terms of solving time and efficiency.

(3) For the power system considering the valve point effect, the constraint processing method is still adopted based on the rate of slight increase in consumption, apportioning the violation of the equality constraint, and then combining with the particle swarm algorithm. Through the analysis of calculation examples of 3 and 6 units, it provides an explanation of its superiority in solving time and efficiency, and at the same time proves the stability of its optimization results. On the whole, this method is very effective in solving the economic dispatch problem considering the valve point effect, and has certain advantages in application.

Keywords: economic dispatch; constraint processing; consumption increase rate; valve point effect; output apportionment; particle swarm optimization

目录

第一章 引 言 1

1.1 研究意义与背景 1

1.2 国内外有功优化研究现状 1

1.2.1常规优化算法 2

1.2.2人工智能算法 2

1.3 本文的主要工作 3

第二章 传统模型及约束处理方法 4

2.1引言 4

2.2电力系统经济调度模型的建立 4

2.2.1目标函数 4

2.2.2约束条件 5

2.3传统约束处理方法 6

2.4粒子群算法 7

2.4.1粒子群算法简介 7

2.4.2粒子群优化算法的数学模型 7

2.4.3粒子群优化算法的求解步骤 8

2.5本章小结 9

第三章 电力系统经济调度的约束处理方法 10

3.1基于机组耗量微增率的约束处理方法 10

3.1.1机组的耗量微增率 10

3.1.2求解步骤和流程图 10

3.1.3求解模型的过程 11

3.2算例分析 14

3.2.1 3机组系统仿真 14

3.2.2 6机组系统仿真 17

3.3本章小结 19

第四章 考虑阀点效应的约束处理方法 20

4.1阀点效应 20

4.2基于粒子群算法的模型求解过程 20

4.3算例分析 22

4.3.1 3机组系统仿真 22

4.3.2 6机组系统仿真 25

4.4本章小结 27

第五章 总结与展望 28

5.1总结 28

5.2展望 28

致谢 30

参考文献 31

1. 引 言
   1. 研究意义与背景

因为现代社会经济发展十分迅速，新时代依此建立的各个领域对电力都十分的依赖，因此电能的需求量也是不断地增加，同时，我国的电网也随之迅速发展。但是煤炭等化石燃料，仍然是作为我国电力工业主要的消耗资源而存在的，因此在消耗和污染物排放方面，我国仍然在世界上位居前列，而且很长时间内难以改变。在如今的社会环境中，随着供电系统中电压水平的不断提高，就会出现诸如负荷率降低、负荷谷峰差的不断增大、各区域电网连接等问题，这类问题的出现，则是对当前的电力系统经济调度问题拉响了警报，也对其提出了更高的要求。为了节约能源，提高电网的经济运行效率，我们需要在提高电压质量和降低损耗等方面下苦功夫。故而，在满足电力系统可靠性和保障电能质量的前提下，可以考虑通过优化分配来节省能源的消耗，以此来提高发电效率，而且在节约系统的运行成本方面，具有极大的实际应用价值。同时，为了保障电网的安全可靠经济运行，在实际中，常常采取配置无功电源合理补偿无功负荷的方法，来维持电力系统的稳定运行和电能质量。

在早期，对于电力系统的经济调度问题，研究人员的重点落在提高系统的经济性上，对其他方面鲜有涉猎，这就是电力系统的经济调度。经典法中经济调度的核心就是等耗量微增率准则，如今已在诸多电力公司调度中心中受到广泛应用。自1965年以来，各个国家发生的许多电网崩溃事故凸显了电力系统安全问题的严重性，世界各地大规模的停电不仅影响了我们的日常生活，更给国家造成不可估量的经济损失。一次大面积的停电造成的损失，甚至比几十年的优化收益还要高，因此，电力系统安全稳定运行的重要性不言而喻。此时，经典法的经济调度已经无法满足电力系统优化运行的需求，在追求经济性的同时，更要考虑安全性的问题。

剩余内容已隐藏，请支付后下载全文，论文总字数：23451字